Re-Examining Racial Disparities in Prostate Cancer Outcomes

VOLUME NUMBER AUGUST J OURNAL OF C LINICAL O NCOLOGY E D I T O R I A L Re-Examining Racial Disparities in Prostate Cancer Outcomes Matthew R. Cooperberg, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, CA See accompanying article on page 2991 Prostate cancer incidence varies greatly across the world, reflect- ing disparate population genetic factors, dietary and other environ- mental influences, and local screening and diagnostic practices. Incidence is high in the United States, where prostate specific antigen (PSA) – based screening is relatively intense; the world age- standardized rate (ASR) for diagnosis in the United States in 2008 was 83.8 per 100,000, corresponding to a 10.8% lifetime risk of diagnosis. The ASR for mortality is 9.7, which is fairly average from the global perspective (although the mortality rate has been falling faster in the United States in the PSA screening era than in any other country). 1,2 Most of the highest ASRs for mortality worldwide are found in western and southern Africa and in regions with large populations of African ancestry, most notably the Caribbean (ASRs up to 61.7). 1,3 Within the United States, mortality ASRs vary more than fivefold, from 9.5 for Asian/Pacific Islanders to 48.2 for African American men. 4 In fact, the mortality rate ratio between African American and white men is higher for prostate cancer than for any other ma- lignancy. 5 African American men tend to present with higher- stage, more aggressive disease than those of other racial/ethnic groups, but tumor characteristics at diagnosis only partially explain the higher observed mortality rates in African American men. 6 Relatively few studies have examined racial variations in outcomes among men who are well-matched for grade, stage, PSA, tumor extent, and other parameters. In the article that accompanies this editorial, Sundi et al 7 report an analysis of men in the Johns Hopkins University prostatectomy cohort who met institutional criteria for low-risk disease and could have been eligible for active surveillance but opted instead for imme- diate surgery. The authors found that, compared with white men, African American men meeting these strict low-risk criteria had higher rates of adverse pathology (defined as Cancer of the Prostate Risk Assessment [Post-Surgical] score ⱖ 3, reflecting higher-grade and/or higher-stage disease) and biochemical recurrence. These re- sults were not only statistically significant but clinically quite substan- tial; among contemporary-era (2004-present) men undergoing extended-pattern biopsy, African American men were more than six times as likely as whites to have adverse pathology on multivari- able analysis. This observation is an important one, with key potential impli- cations for both clinical practice and future research. Whereas the independent association between race and prostate cancer outcomes Journal of Clinical Oncology, Vol 31, No 24 (August 20), 2013: pp 2979-2980 has not been demonstrated consistently in previous studies—as re- viewed by Sundi et al 7 —the observation of disparate results specifi- cally for those with low clinical risk has been made at least once previously. A study from the Shared Equal Access Regional Cancer Hospital registry, representing multiple Veterans’ Affairs centers with a relatively high proportion of African American men, found that among men with low-risk characteristics at time of surgery, African American men were substantially more likely to experience biochem- ical recurrence than others. 8 Assuming, then, that the observations by Sundi et al 7 may repre- sent a true phenomenon, two principal questions raised in the article are particularly salient. First, should African American men be consid- ered ineligible for active surveillance on the basis of a higher rate of apparent undersampling? The answer is of course no, but these higher risks should be included in the information presented during decision-making counseling sessions with African American men newly diagnosed with clinically low-risk prostate cancer. The study authors suggest that existing risk instruments should be recalibrated for African American men, or that race/ethnicity should be included as an additional parameter in multivariable models. In fact, in develop- ment studies for pretreatment multivariable predictive models such as the Kattan nomogram and the Cancer of the Prostate Risk Assessment score, race was expressly examined but was not an independent pre- dictor of outcome after adjustment for PSA, Gleason grade, stage, and biopsy details. 9,10 Others have shown that models predicting post- prostatectomy outcomes perform with comparable accuracy and cal- ibration between African American and white men. 11 Another possibility might be a relatively intense approach to surveillance for African American men, including, for example, earlier repeat biopsy and/or more use of adjunct imaging, such as multipara- metric magnetic resonance imaging. Emerging biomarkers may be valuable for these men as well, although most such markers to date have been validated only in populations that are overwhelmingly white. Of note, recent studies have identified different expression levels of prostate cancer–related genes between African American and white men, even after adjustment for clinical parameters. 12,13 Finally, a lower threshold for intervention might be appropriate for African American men initially opting for active surveillance. Decisions for intervention should also be informed by the fact that from a psycho- logical standpoint, African American men experience substantially greater traumatic stress symptoms after prostate cancer diagnosis than © 2013 by American Society of Clinical Oncology Information downloaded from jco.ascopubs.org and provided by at CONS CALIFORNIA DIG LIB on July 14, 2016 from Copyright © 2013 American Society of Clinical Oncology. All rights reserved.